Monday, September 30, 2024

Untitled


(from pg)


I mean, it's curious how this pragmatic, worldly advice mirrors the more spiritual perspective.

Decisions are made the moment you know what you will do. Decision making is a process of knowing yourself. To know yourself, you must know the world, because you are connected and immersed in it.

The spiritual aspect is that we all basically know what we want. The root cause of indecision is not due to lack of knowledge per se, but fear. Fear makes us unable to decide. Spiritual knowledge is always with us. The answer is clear -- all paths are valid, and free will is a gift to us to do whatever we want. With freedom comes constraint, as things come in pairs. One we have chosen we cannot choose the other paths. At least not simultaneously.

The optimizer says we must choose the best. Whatever the objective function, the optimizer purports that there is a best choice, and thus all else are invalid.

The fallacy of the optimizer mindset is that we do not know the objective. The optimizer presupposes that we know -- and in many social contexts, that is partially true (for example, businesses optimize to make money).

But in life, we do not know the objective function. We can only take one step at a time, look back, and realize in retrospect what the objective actually was (if there was one that we understood). (This is, basically, what Steve Jobs tried to say in his famous Stanford Commencement Address.)

For those uninitiated, spiritual knowledge is very subtle, and often goes unnoticed. It must be so, because here free will reigns supreme. Spiritual forces cannot dominate any decision we make here by free will. If we freely decide to fear, basically fear that we will have made a wrong decision, then we will not know the decision to make.

Fear is always fear of death. If we could relive and retry, will we really fear taking a wrong turn?

Whenever we fear, we cannot know. To know, we cannot fear.

From the spiritual perspective, *learning*, whether through studying texts or through lived experience, is merely a worldly ritual to kind of crystalize the information. It's kind of optional once you realize that knowledge (in gnosticism it is called "gnosis") can be obtained through other means. One crucial step is of course eliminating fear.

When fear is truly dispelled, the ego disbands as well. What remains is knowledge that does not have an apparent source.

But we chose to be human, which "unfortunately" requires having at least a bit of ego. So the practical method of actually going out and acquiring knowledge is usually the more effective way to go about doing things.

Thursday, September 26, 2024

Quick Notes (S)

- Emulation is the highest form of resonance known to us.

- The "middle way" is probably code for "not yes, not no, not not yes, not not no" -- why? because it really depends on which layer you're talking to. when doing something has opposite effects on different layers, answering the yes/no question is impossible, and you must assert so. Explaining the different layers is a bit tough, so probably they just left it as an exercise for the reader to connect to spirit and understand on that level instead...

Quick notes

- In UK it is common to inhale the smoke/vape of the pedestrian walking in front of you
- Exits are universally called "way out" for some reason.
- London is grand and splendeur but when compared with places with history, the London version reeks of 暴發戶
- Best places: Mull of Galloway Lighthouse, Fountain Abbey, Oban (the town)
- 包場 experiences - 免費嘅嘢通常最貴
- Clydeside distillery - the new make spirit tastes like 醬香茅台
- Stone fences vs wooden stake and steel wires, which is more effective?
- 耐心處理問題,學到酒店業嘅黑幕 OYO
- London 大塞車, google map 老點我走旅遊區,喺西敏寺外塞到 hihi
- google map 教我上公路即刻喺 exit 出⋯然後入返去。公路大塞車。慳咗我諗起碼十分鐘。
- 乾隆啲香爐好似鼎..
- 西敏寺入面感覺上係啲人「周圍咁隨便安葬」搞到入面啲佈置感覺好亂,一啲都冇簡潔嘅美感
- 英國啲希臘式建築基本上係 希臘抄古埃及、羅馬抄希臘、英國抄羅馬.... 見得越多越覺得件事好膠,尤其係啲 19 世紀先起嘅希臘式建築,真係有種港女買LV覺得自己好high class嘅感覺。
- 啲 gothic architecture 反而順眼好多
- 去咗食 afternoon tea,見到兩位女士食咗一個 cake 就停咗手 (佢個 tea set 係有每人四個三文治嘅,所以我理解,但相機食自己唔食呢家嘢...)
- The clock is a wonderful technology for its time. Absolutely game changing for so many reasons. The fact that Britain has so many clocks everywhere is just a reminder where its glory days lie.
- 喺 st paul's cathedral 聽咗陣 pipe organ 幾好聽
- 某酒店嘅 3xx 同 4xx 號房喺同一層,然後 1xx 號房喺另一棟大廈...
- 有啲泊車超先進, QR code + Apple Pay;有啲只收銀仔 (紙都唔收) 但最低消費五磅...
- Roman baths 底下啲石嘅大細同擺位有埃及 feel。都幾大塊石頭。有可能係更早前已經有建築 (羅馬啲石一般冇咁大)
- 另外原來 Roman baths 係有泉水湧出 (有泡泡𠻹)。但個大池仍然係好似一潭死水咁
- Carlisle Castle 早入場,見到個工作人員走出門口放個 "open" 牌 (背面係 "closed" 嚇咗我哋一跳). 感覺包場,睇嘅嘢同倫教嘅城堡(Eg.  Tower of London)差唔多
- Caerlaverock Castle 又係包場嘅感覺。冇乜人,好似獨遊遺跡咁。不過 Caerlaverock 真係好細,話係一座用石頭起嘅大屋就差唔多。 (但有真 moat!!)
- McDonalds saved us from starving more than once
- I ate a bit of grass from either stonehenge or castlerigg stone circle (probably stonehenge though) and drank a bit of water from Bath's spring.

Thursday, September 5, 2024

Meta killed OpenAI

In a very real way, Mark Zuckerberg killed OpenAI by releasing Llama (2,3) as a (mostly) free open weights model.

Before that move, it was plausible that those who controlled Very Large models could form some kind of oligopoly and charge the models not at inference compute cost, but by amortizing the development/training costs of the model itself.

But now,  everyone is forced to serve majority of their models at around inference cost, because if you don't keep the price low enough, people will just use Llama3 70B instead, which is pretty much indistinguishable from SOTA models.

Releasing SOTA models for free after spending billions of dollars on training is an extreme case of predatory price cutting. It's wonderful and horrifying at the same time.

Monday, September 2, 2024

carb

近排發現自己肥咗唔少 (相對)

唯有減下 carb

一減就發現嘩搞唔撚掂

睇嚟真係要清一排調整返身體狀況


Sunday, September 1, 2024

$

由於識咗 PK Ming 而佢日日喺臉書share靚女相 (啲女係靚嘅),見到 Copy 呢單花生

https://www.facebook.com/share/p/DspKzZiHzFNq8Fg4/ 

講到咁委屈,去到尾都係嫌條仔窮。

咁當然人最緊要首先要照顧好自己先,但作為男性,啲女話唔介意你窮,真係咪撚鳩柒信。 :0)

The Big Ponzi

It's obvious that modern society is run as a Ponzi scheme.

At least the kind that attributes seniority based on age or experience.

If people can generally earn a better wage by being senior, and being senior simply implies having more experience doing the same thing.... then basically it means as people grow older they can expect to earn more money by doing pretty much the same thing (or sometimes even less).

And how does society pay for this? It seems to me they simply exploit the younger workforce with the promise (often fulfilled) that they, in time, will have the same benefits just for consistently showing up.

It used to be that the scheme also run under the assumption that the population is increasing, so there are always more junior positions for a senior position to "feed" upon, but that part of the equation is being disrupted.

But it's still sufficiently close to a Ponzi scheme.

There are other Ponzi schemes like social welfare (esp. for the elderly) that is paid for by a rising population.  There mere idea that the economy is going to take a hit due to falling population is a hint that there's something Ponzi in the existing assumptions.  No regulated fund dares tell investors that they're taking a loss because the number of investors have dropped. Yet economists keep saying that shit with a straight face.

Disgusting.

And all of that is obvious. Everyone with a sane mind knows there are Ponzi schemes running everywhere (as long as there's money involved and it's government sanctioned, it's sus. Try this heuristic and be amazed.)

My point though, is that Ponzi schemes make people "lazy" and lose ambition. It incentivizes people to fight to maintain the status quo.

And that's bad for economies actually.

We're seeing this play out in many developed economies. They get to a stage of advanced development, then settle into some kind of Ponzi scheme (most notably -- retirement pensions, social welfare, or some kind of social contract where employers don't ever fire employees even if they're underperforming.) People who know they aren't pulling their weight are afraid to lose their senior status in the Ponzi schemes they participated in, so they fight tooth and nail to maintain the status quo.

And maintaining the status quo means stifling development and improvement.

This isn't like a social justice critique of how unfair things are and how boomers have it easy. (Although you can take it as such.) This is a reflection after thinking deeply about why some economies are doomed to decline in the coming decades, and why some economies are growing.

In particular, this thesis can in a sense explain the continued growth of USA after it had attained developed country status -- because its social net is shit, employees have little protection and are generally hired and fired at will, and labor unions don't affect the industries that are growing (tech). People are still motivated by poverty, and they don't have a lot of proud traditions to defend relatively speaking.

Economies that are more equitable seem to take the hit more badly. See even Japan, which didn't even experience the boom that long (about a decade or two between 1980-1990?)... and they just fizzled out. For some reason in the 21st century they just stopped the large scale innovation that they were known for in the 20th century. Where they used to be known as the high tech nation, these days (gaijin) friends working in Japan complain that they're still fixated on *fax machines*. There's a limit to "if it ain't broke don't fix it", and I guess this is it.

So if you're investing long term -- this is the kind of forward perspective you should be adopting. Too many people simply look at the past and project -- it might work for short term bets, but it never works long term.